The present invention is related to message processing, and, more specifically to processing a message queue utilizing a software or computer agent.
The term “agent” describes a software abstraction, an idea, or a concept and provides a convenient and powerful way to describe a complex software entity that is capable of acting with a certain degree of autonomy in order to accomplish tasks on behalf of its user. But unlike objects, which are defined in terms of methods and attributes, an agent is defined in terms of its behavior. An autonomous software agent is a stand-alone computer program that performs tasks on behalf of a controlling entity, but without direct, continuous supervision. The autonomous agent may utilize network resources and services, but works relatively independently of other software agents that may be a part of a multi-agent system.
Since agents are well suited to include their required resources in their description, they can be designed to be very loosely coupled and it becomes easy to have them executed as independent threads and on distributed processors. Thus they become distributed agents and the considerations of distributed computing apply. Agent code is particularly easy to implement in a distributed fashion and scales well.
With distributed agent systems, it is desirable for an agent to process only one message at a time so that data integrity is maintained and race conditions are avoided. In such a situation, problems arise when, during the processing of a message, blocking occurs because of another message sent to another agent causing the agent to become unresponsive for some period of time. This period can sometimes be in the seconds or even minutes. Therefore, what is needed is a solution that overcomes these problems and limitations.